Topic: Digital Servo's newbee questions

Having just got my DX7 I now have some nice digital servos as well. I've read (elsewhere) that using Y-leads isnt recommended. I havent used digital servos before and before I deploy them in a model I want to know if there are any best practices I need to be aware of. The DX7 manual doesnt give any information other than dont use threadlock on the arm screw.

Over to you gents!!

I don't understand why you shouldn't use "y" leads with a Dx7. But I have read somewhere that ferrite rings may cause problems if used in conjunction with extension leads, but nothing about "y" leads.

Don't use "y" leads myself. The software in my brand of radio renders them redundant.

I see several people have had problems using the supplied DX7 4-cell pack and digital servos.  Seems the Rx will reset and go awol for a short while when spiked down to around 3.2v which can be caused by current hungry servos (digital). Advice generally seems to be use low impedance batteries, avoid long battery cables, and possibly use a 5-cell pack.  Not a problem as such, just a little more care needed.

I have no personal experience to recount as I'm still on 35MHz.

Hi Glowfly

Multiplex now supply a plug in filter to address the very problem you've just mentioned! The cost is £4.99

You should NOT use filters such as ferrites with a Spektum 2.4Ghz set. The manual specifically recommends aginst this, and it is totally useless in a 2.4 Set.
As for the "supplied 4 cell pack " - the UK sets do NOT get a receiver pack, just the 4 digital servos as mentioned, and the AR7000 rcvr. I use a 5.5V regulator / switch on my Spitfire, and 2 of the digital servos , on a Y lead configuration ( but actually hard wired via an ashlock ) and have had no issues. There have been some reports of the sets misbehaving if the receiver pack drops to a low V...but so will any recvr complain when its at its low threshod V.

I would think the don't use a Y lead comes from the fact that digital servos a relatively power hungry devices, feeding 2 from a single outlet is asking for trouble. If you must use a single channel to feed 2 servos I would be very tempted to supply them with power seperately i.e. not down the Y lead.

Regards - J

I've read with an element of mirth the problems pertaining to incorrect use of the supplied RX pack in US sets. Up to reading those posts I wasnt aware that digital servos were more power hungry than normal servos. Labelling a post 'DX7 problems' when the actual problem relates to incorrect battery sizing is a bit off IMO.
So far none of what I've read in the supplied responses here leads me to believe that I should use them in any way different to my use of standard servos, as long as I remember to use adequately sized RX packs. I am curious however why a ferrite ring/torroid would have a detrimental effect on servo operation. Surely the point of a torroid is to reduce induced RF interference on long cable runs within the servo wiring loom itself, due not only to RF from the TX but the general ambient RF environment.

Keep the responses coming please!!

I am not sure why Spektrum advise against the use of ferrites etc, but as for a reason as to NOT use them, how about .....It is a complete waste of time, money, and carries a weight penalty. The frequency of any generated RF noise in the servo leads is WAY below that of the radio frequency deployed in 2.4G technology.
 I think also, that the comment made about a reason not to Y lead them on a single rcvr output ws more a concern of the capacity of the receiver pins /circuit board to handle the current involved. Not my  input incidentally....just my interpretation.

There have been rare occasions when Hitec digitals on a Y lead would not work.  Some of their servos had a very high threshold for recognising the difference between high and low signal, the current draw of both digis on one data wire dropped it below their high threshold so both servos saw a permanently low signal.

I take it that I dont have to go all digital on a given model and that a mix of servos is OK?

No you dont -yes it is

Tremendous.

So in terms of fixed wing IC powered flight, what channels are they(digi's) normally used on, elevators, ailerons?

Depends on your needs. Only real advantage over a good analogue is that they are quicker, and self centre better. You will doubtless hear a constant buzzing from them, as they try much harder to centre accurately, and often just the weight of the control surface itself will cause a buzz....quite normal.

In terms of RX pack voltage, am I better off going 6V with digi's?

Generally yes ACE, but you'd need to check the servo manufacturers specs carefully as some are designed for 4.8v operation only.

And of course remember that current will INCREASE if using a 5 cell pack, so duration will be LESS - bigger pack (capacity wise ) needed

As for the "supplied 4 cell pack " - the UK sets do NOT get a receiver pack

Ah, it must be references on US sites that I remember then since they DO get a Rx pack.

There have been some reports of the sets misbehaving if the receiver pack drops to a low V...but so will any recvr complain when its at its low threshod V.

The difference here is that the Receivers containing a MPU typically take MUCH longer (sometimes by orders of magnitude) to start working again compared to the simple stateless Rx designs we were previously used to.  It applies not just to Spektrum product but several other MCU based products. I believe it will become more of an issue in the future unless clear guide lines are given or a slightly different power design is used.
Industrial equipment designers have understood this for many years and usually take greater care in the power distribution source impedance from lead resistance and inductance than typically seen in our  models.

Industrial equipment designers have understood this for many years and usually take greater care in the power distribution source impedance from lead resistance and inductance than typically seen in our  models.

Sorry ? Wassat mean then ?
 

Sorry ? Wassat mean then ?
 

They use thick cables, keep them short and heavy duty switch gear.

Regards - J

Ah ! That I can understand     

Just encase you want to use these digital on other models please be aware that digitals can interfer with PPM RX's.

I've had very bad interferance when I have tried JR 8511's & 8411's on Futaba PPM RX's but they work perfectly on Futaba PCM. This is not myth and I can reproduce it on the ground every time.

I would also suggest you invest in a good battery tester as they will drain your battery faster. How much depends on what servos, how many, your battery and your flying/model.

By all accounts though they seem to be very nice servos and I would not want to put you off, just offer experiance I have had.

Just encase you want to use these digital on other models please be aware that digitals can interfer with PPM RX's.

I've had very bad interferance when I have tried JR 8511's & 8411's on Futaba PPM RX's but they work perfectly on Futaba PCM. This is not myth and I can reproduce it on the ground every time.

I would also suggest you invest in a good battery tester as they will drain your battery faster. How much depends on what servos, how many, your battery and your flying/model.

By all accounts though they seem to be very nice servos and I would not want to put you off, just offer experiance I have had.

Is that all digitals and all PPM receivers ? If so, what can be done about it ?

Sizzling, might there may be some other problem you are encountering?  Digi servos should not interfere with any radio.   I use an 8411 (along with many other digi servos) and all with PPM.  If the problem disappears with PCM one possibility is that one of your servos is faulty and causing rf which is being masked by the pcm popping in and out of hold mode too quickly for you to see it.  Another possibility is perhaps the current handling or buffering of spurious signals from the servo leads in the output section of that pcm rx is of better quality than in those particular ppm rx you are using so a faulty servo is causing the problem and it is coincidence that the pcm rx is blocking it better.

The frequency of any generated RF noise in the servo leads is WAY below that of the radio frequency deployed in 2.4G technology.

I wouldn't be as confident in the assertion that servo amplifiers are incapable of emitting in the GHz region, but that's really not the point. Just because it's a 2.4GHz receiver it doesn't make it imune to signals at other frequencies. Television signals are at frequencies way away from out band, but a 35MHz transmitter will often interfere with a TV set because 35MHz is quite close to one of the IFs used internally in the TV reciever. Numerous UHF (and higher) systems on aircraft have proven embarrasingly susceptable to interference in the VHF (and even long-wave) bands. The Tornado was at one time prone to suffering some severe problems if it flew too close to local radio station transmitters.

The law of unintended consequences is ignored at one's peril. A few years ago a certain american manufacturer started selling its speed controllers in the UK, but didn't bother to check whether an item which worked fine with 72MHz equipment was actually compatible with 35MHz systems. It wasn't, and the reputational loss cost FAR more than the few thousand dollars worth of redesign and testing...

PDR

I only have two digital servos, an 8511 and an 8411. When I upgraded my Funtana 140 rudder from a non digital to the 8411 I instantly started getting interferance with the model on the table. It had an 8ch Futaba Dual Con RX, it had me going for a while as I thought I had upset the RX in the change of servo. By luck I had a spare PCM to hand so I tried it and bingo interferance gone. This was getting a bit strange so I put the PPM back and interferance was back again....hmmmmmm so I unplugged the rudder servo and instantly the interferance stopped. Further testing on another PPM RX I got out of a model that had been flying with out a glitch for about a year and exactly the same, as soon as I plugged in the digital the whole system was started glitching.

Since then I have bought an 8511 as the 8411 was such a good servo and made such an improvement in my Funtana I though my Yak deservered similar quality. I did however try it on another PPM RX to see if it did the same and guess what, yes it did.

Forgot to mention, all my RX's are Futaba and all other servos were Futaba so this limits my tests.

The reason I fealt safe flying was after a good talk with my LMS manager who is a good flyer and I know will tell me his honest opinion he instantly admited he had had the same issue but on his Heli's and now only uses PCM, he also said he had the problem on other digitals too. Well after some range testing I was convinced and touch wood every thing has worked flawlessly since on PCM.

By luck I had a spare PCM to hand so I tried it and bingo interferance gone. This was getting a bit strange so I put the PPM back and interferance was back again....hmmmmmm so I unplugged the rudder servo and instantly the interferance stopped. 

The reason I fealt safe flying was after a good talk with my LMS manager who is a good flyer and I know will tell me his honest opinion he instantly admited he had had the same issue but on his Heli's and now only uses PCM, he also said he had the problem on other digitals too. Well after some range testing I was convinced and touch wood every thing has worked flawlessly since on PCM.

Just remember that PCM doesn't cure interference it hides it and there is a subtle difference that can have very serious consequences. I personally would have tried to cure the interference more with ferrite rings / opto isolator / cable routing etc.

FWIW I have Hitec digitals on quite a few models, no interference at all on anything including 2.4GHz 

Regards - J

FWIW I have Hitec digitals on quite a few models, no interference at all on anything including 2.4GHz 

Does that include Futaba PPM? As this is what my experiance is based on.

I really wouldn't want to go down the optio isolator route but the ferite ring sounds simple enough, is it something my LMS should stock? Is it just a ring or an extension lead with the ring on it. The RX's in both cases are only about 4" or 5" from the servo so not much need for an extesion lead.

Multiplex now supply a plug in filter to address the very problem you've just mentioned! The cost is £4.99

What is it Brian?  A battery bypass capacitor or what?

I gather Spektrum offer a plug-in capacitor that goes into an unused servo connector to help minimise battery drop-out spikes.  Seems the wrong way to solve things to me since the Rx is probably more sensitive to the issue than most servos. I DO know that there have been occasional problems too with digital servos suffering from an internal reset and subsequent longish recovery with long/high resistance servo leads.

It all makes the idea of a Powerbox type of product that can provide high current power distribution to the servos more attractive in larger models.  Seems like such a unit could easily provide a separate stable supply to the Rx as well. Perhaps it already does?

So far in my larger models I just use two heavy duty batteries and switches and thick cable/extensions and haven't had a problem with IPD receivers. But I don't use 8411 type servos though.

I wonder whether using 2 cell A123 LiIon type batteries might be useful as a source of low impedance Rx batteries without needing a regulator. I suppose it would be about 6.3v under load and a very stiff supply.  Possibly some servos might not like the extra voltage although the Spektrum receivers would be fine with it.

Sorry ? Wassat mean then ?

Sorry about that! Johnb80 knows what I meant! 

Does that include Futaba PPM? As this is what my experiance is based on.

Yes

I wonder whether using 2 cell A123 LiIon type batteries might be useful as a source of low impedance Rx batteries without needing a regulator.

I changed my Sabre over to Saphions a while ago but am changing back to nicads because that extra bit of voltage makes all the servos chatter excitedly when it is fully charged.  I had banked on the battery backer knocking out most of the extra bit of voltage the Saphions have (nominally 6.4V compared to 6V) but the peak voltage of a freshly charged pack is just too much.

I changed my Sabre over to Saphions a while ago but am changing back to nicads because that extra bit of voltage makes all the servos chatter excitedly when it is fully charged.  I had banked on the battery backer knocking out most of the extra bit of voltage the Saphions have (nominally 6.4V compared to 6V) but the peak voltage of a freshly charged pack is just too much.

Now that's disappointing news.  I thought it might turn out to be a good solution.  What servos are you using on the Sabre Harry?  Does the chatter completely disappear on a freshly charged NiMh pack?  When it occurs with the Saphions is it just a more severe form of the digital buzz we're used to - but with significant visible movement?

Thanks.

It's a mixture of JR digi, JR analogue, Futaba analogue.  Its a buzzing bordering on fluttering, that I don't get with a freshly charged 6volt nicad.  The servos are making it clear that they are not happy.  They settle down after a couple of flights and the top peak of the voltage has gone.  But remember that I am using a backer than removes 0.2 to 0.4 volts, if you feed them the raw voltage from the saphions then the problem will be worse.  Perhaps they are suitable for feeding a regulator with a 4.8V output but I don't like adding yet more possible failure points and my rx demands 6 volts for its internal backer anyway.
I got around to pulling out the Saphions and replacing them with nicads this evening so I feel a lot happier now.

Do you find your freshly charged 5 cell Nicad really that much less than a 2 cell Saphion?  I assume you terminated the Saphion charge at 6.6v rather than leaving a LiPo type charger to completion?
I find my 5 cell packs measure around 6.4v after charging.

Oh well, maybe I'll stick to the batteries I know best!

I set charger to lithium ion rather than lipo for the saphions, believe this reduces it by 0.1v/cell compared to lipo.

I apologise to everyone for hijacking this thread, but while I have your attention Harry...

I understood the A123 batteries unlike traditional LiIon only need to be charged to 3.6v per cell for full capacity. They will tolerate being charged to 4.2v/4.1v in exchange for slightly shorter life. Also I thought there was negligible capacity gain by charging beyond 3.6v/cell.  Sorry for banging on about this but I just wanted to be sure your chatter problem (which I'd like to avoid) isn't just caused by an initial overcharge which is shed after the first flight.

OK I'll stop now, I promise. 

I think I might have posted before about not using a Y-lead with the digital servo's I use.  But that is begause the are high power 15kg ones which draw some serious current, again as has been said in this thread sub c batteries and a quality rx were also needed to handle the current.

However with these standard size digitals I would imagine normal standard radio gear would be fine.   Overlander do some packs which are shorter but fatter than AA's and I'm told that they can hold voltage under load a bit better than AA's.  But personally a good quality GP or sanyo 4 or 5 cell pack of over 1500mAh is what I would use.  You can get them for around 18 - 20 each maybe less.

CM

I understood the A123 batteries unlike traditional LiIon only need to be charged to 3.6v per cell for full capacity. They will tolerate being charged to 4.2v/4.1v in exchange for slightly shorter life. Also I thought there was negligible capacity gain by charging beyond 3.6v/cell.  Sorry for banging on about this but I just wanted to be sure your chatter problem (which I'd like to avoid) isn't just caused by an initial overcharge which is shed after the first flight.

It took a flight or two to drop out the excess voltage chatter.  Charging to a lower voltage might do the trick but has some snags: I can’t put it on and walk away, I have to keep watching it as there is no option in my chargers afaik to set 3.6V as the actual cut off; when not fully charged the voltage is dependent on the current and is therefore variable for example 3.6V at 2 amps is a partially charged perhaps almost flat battery, 3.6V at 100mA is nearly full.  So to charge it to 3.6V at some low current but do it in a short time at the field means starting at some high current then stopping the charger and dropping the current and repeating until 3.6V at a low current is achieved.  I guess I could do it without too much hassle if I really want to, since I do a load test on each battery before every flight so there is no danger of starting off with a low battery.
I am so used to charging rx in nicad mode that the danger exists of forgetting to change mode (I did once but realised after a few minutes) and then potentially overcharging since there will be no peak to stop the charger.

I can see your problem regarding convenient charging, thanks for answering Harry.

I was thinking of using a 5 cell sub C pack of over 2000mah for the RX after reading through all the replies, but I cant find any cell stats for discharge rates for the (GP) cells overlander use. I already have a couple of 5 cell 2000mah nimh AA pack in my Sea Fury (8 std servos and 1 retract on a seperate pack) and by the looks of the draw when I recharge the packs I'm hardly using them, as I've never had to put more than 500ma in either, even after a full days flying. Maybe I should give them a go, but in this game you only get 1 chance so an over engineered solution may well be prudent.

Ace, sub-C packs will happily cope with the current draw of rx and servos unless you are using lots and lots of digis in which case other parts such as thin wires and pins and switches are more likely to give problems, and the fact you are using 5 cell packs should prevent any voltage drop on load from getting below the critical voltage somewhere under 4V.  My Sabre using 11 channels with 8 servos (1 digi), two electronic valves and an ecu works very happily initially with 2 6V nicad 1700Mah then changed to 2 1500mah saphion and have now changed to 2 1100mah nicad.  I load test before every flight and recharge if required during the afternoon, the batteries have not been a problem and the mAh drawn is not a worrying factor.

What I would suggest is plugging a good battery checker into your RX on the ground and just giving the sticks a good waggle. On one of my models with 4 non digital 9.5kg servos and 1 11kg digital and a throttle servo I had a 5 cell 1800 mha nimh AA pack that had been peak charged and was no more than 20 cycles old, on waggling the surfaces the checker would drop into the red, but on changing to a 2000 mha NiCad all the stick waggling couldn't get the pack to move off top green light.

This I beleive is why a lot of the high power digitals clearly state on the instructions not to use nimh.

 

This I beleive is why a lot of the high power digitals clearly state on the instructions not to use nimh.

Mistakenly in my view. It may have been the case when NiMh packs were first introduced, but it's no longer necessarily true that NiMh have higher internal impedance than NiCd packs. Sadly there's no real substitute for looking up the actual specification for the cells you plan to use if low resistance is important.  The very high capacity AA NiMh cells for sale on the high street are a case in point. They may offer that capacity at low currents but seem unable to supply the high current peaks to thirsty servos without severe voltage drop.

Both my bigger models with current hungry servos have onboard checkers that are easily viewable and at a wiggle of the sticks I can easily see any power drops. Which ever type of battery you use, if the model needs power hungry servos that are expensive I think it deserves a good onboard checker at a few extra quid for peace of mind.

 

That's a very good point sizzling, however i would just point out to any users of Multiplex 12 xtal or 9 or 12 synth IPD rx who use the internal battery backer that an on-board battery monitor should not be plugged into the rx as it will give a false low reading due to the voltage lost in the battery backer.
I expect the same is true if you use something like the SM services battery backer.
You could use an on-board monitor but would have to solder on an additional lead or add in a Y lead from the switch output and plug the monitor into that, then it will not suffer the battery backer's voltage loss.

As this thread seems to have digressed slightly into the area of which battery type is best for digitals /high drain use, could I be cheeky and ask for any comment to an older thread of mine which seemed to get sort of lost :(
http://www.rcmf.co.uk/4um/index.php/topic,39653.msg424498.html#msg424498

As this thread seems to have digressed slightly into the area of which battery type is best for digitals /high drain use, could I be cheeky and ask for any comment to an older thread of mine which seemed to get sort of lost :(
http://www.rcmf.co.uk/4um/index.php/topic,39653.msg424498.html#msg424498

Errm - you want us to think of a reason NOT to use your LiPos?  Maybe the lack of reply means no-one could think of a reason not to? 

As long as your particular packs and Powerbox BEC can supply your peak servo load then I can't see any reason not to use them. Even if they are modest in discharge rate (I don't know the particular pack you have) I would still expect them to hold up fairly well for servo usage. Provided you check their state of charge between flights to be sure you have enough charge left they should be fine.

OK thanks for the input, one never knows when a thread stays dead - is it in the wrong section etc ?
one is grateful

A few posts ago we were discussing the current drawn and the need for large capacity etc.  Thought you might like to know the results of refitting my Sabre with 2 x 1100mah 5 cell nicads.  They have now been through 2 flying sessions each of 8 flights.  The first session of 8 flights was off the initial forming charge of 14 hrs at 0.1C, and the second session of 8 flights off a peak charge.  The nicads have to power 8 servos inc one JR8411 digi which is always buzzing under load due to an unbalanced all-moving tail, and 2 electronic valves which take quite a bit of current when wheels and doors are in the up position i.e most of the flight.  I didn't recharge during the sessions and at the end the SM services tester showed under load that each battery is still giving 6.2v under load, and at recharge each took 450mAh, total 900mAh.  That was for a total radio switched on time of 1hr 15 minutes.
This doesn't tell us anything about how well those particular batteries will hold up their momentary voltage under the most severe current draw the Sabre can demand, but it does show that at 5 cell currents, 2200mAh is more than enough for a long day's flying of this sort of model and flying style with lots of servos - assuming the batteries are in good condition and of good quality - these are new batteries and I won't let a battery that doesn't have a Sanyo label into my models any more.

Thanks MPX...very useful info. I am thinking my 2 x 900 m/a LiPos into my didgiswitch will be more than sufficent then. I tend to only fly this plane a couple of times each session, with an average flight time of around 15mins. I shall still just check the loaded voltage, and current drawn while all servos are twiddled